Data Collection Procedures

The duration of the course was six weeks, with two sessions per week and each session lasting for two hours and 45 minutes, for a total of 12 sessions per level and a total of

33

28 instructed language teaching hours (excluding the final session where the students had their final exam). The researcher conducted the study in three classes. Each class consisted of 15-19 students, which made a total number of 49 participants. However, some of the

participants were excluded for different reasons such as dropping the course before

completion or participating in only one or two of the study's three stages. One student was a non-technology user and another one was not Egyptian. The total number of participants who contributed in the current study is 45.

Learners had two quizzes throughout the course where their listening and speaking skills were assessed. They listened to a conversation, took notes, and answered some true or false and MCQs, and then they did a role-play where they spoke about a real-life situation. In the final exam, learners had similar types of questions in addition to a reading passage and one or more writing prompts to write a short paragraph about. Although reading and writing were not assessed in the quizzes, learners practiced the two skills in addition to listening and speaking during the six weeks every time they covered something new.

Both in-class discussions took place during the regular classroom schedule, where a topic was opened to learners for general discussion. The two topics were based on the lessons already covered with the teacher, so that the students would have adequate vocabulary to participate. The first in-class discussion was held at the end of the second week, after explaining the first unit and having learners solved several exercises about it. The e-

discussion took place at each of the learner's homes, work places, or lab, depending on which location best suited them, since the e-discussions were asynchronous. The students had 48 hours to participate in the online discussion. Learners were expected to participate and speak about the suggested topic from their point of view as well as ask for clarifications if needed and comment on a colleague's post, much in the way they did in the in-class discussions. Instructions were given to learners before holding the e-discussion with regards to length,

34

using only the English language, and what the teacher expected from them. The learners voted on which website to use for holding the e-discussion-- a blog, Facebook, or any of the current academic websites-- to make sure that all of the learners could use it. The learners chose Facebook as their means of communication. Afterwards, a short demo took place in class to make sure all learners knew how to access the e-discussion page. The e-discussion took place in week four right after covering the second unit. The second in-class discussion, however, was at the beginning of week six.

Measures were taken to ensure that the level of FLA was not affected by any kind of anxiety or demotivation because of the topic itself. First, the researcher chose topics that were easy and reflected the learners' everyday interests and activities. In addition, all participants of all classes received an equal amount of preparatory vocabulary and background

information about the topic before participating in the discussion.

3.5 Piloting

The main aim of piloting the study was to ensure that the instruments answered the investigator's research questions. Another aim was to check if any modifications needed to take place with regards to the instruments used before the actual data collection.

The participants who participated in the pilot study were adult Egyptian EFL learners at a language program in a private university in Cairo where English is taught as a foreign

language. The level of this class was a bit higher than the actual level of the participants in this study. However, the age range and learning conditions were similar. A total of 13 learners participated, five females and eight males, and all of them were at an intermediate level. An observation took place among all learners; however, only five filled out the questionnaire with one of these five participating in the interview.

35

The researcher piloted the three instruments with no changes to the above-mentioned plans, except for omitting the e-discussion. As a result, the recorded interview lasted for only 10 minutes, since some questions about the e-discussion were not asked to this participant. Since only five participants filled out the questionnaire, the data analysis for the three instruments was qualitative. Further, these instruments were supposed to be administered twice, before and after the e-discussions. However, since the participants who were part of the piloting did not have e-discussions, the instruments were given only once. The main themes found in the three instruments were communication apprehension, fear of negative evaluation, and test anxiety.

Piloting the instruments was very vital for the researcher; the process helped to determine the appropriateness of the current study’s instruments. Learners indicated that the instructions and language used after translating the questionnaire were clear. However, one item in the FLCAS was confusing for more than one participant when translated into Arabic. The investigator amended the item to avoid any confusion. A suggested alternative of "I don't understand why some people get so upset over foreign language classes" was: "I get

astonished why some people get so upset over foreign language classes". Further, all items used in the FLCAS did not focus on anxiety related to other skills, where communication apprehension exists in much the same way that speaking apprehension does. Consequently, adapted items from other questionnaires were added to address other skills rather than speaking only.

3.6 Data analysis

Data gathered from the observation checklist was used to answer the investigator's first research question, which addressed the effect of e-discussions on learners' FLA levels. Collected data was categorized and coded, to evaluate the frequency of FLA signs before and

36

after the e-discussions. The first section in the observation checklist was analyzed quantitatively by counting the frequency of learners’ FLA signs before and after the e- discussion. In the second section, data was analyzed quantitatively and qualitatively through the Text Content Analysis Tool http://www.usingenglish.com/resources/text-statistics.php. This tool not only compared the number of words pre- and post-online intervention, but it also measured utterance length and complexity of vocabulary and structures used.

Data gathered from the FLCAS was analyzed quantitatively, using a paired sample t- test, to compare the FLA levels before and after the e-discussion. In addition, the FLCAS data was also analyzed using a non-parametric Mann-Whitney test, to determine if gender was a variable in this study, especially that the gender distribution was imbalanced. Thus, the questionnaire answered the first and the third research questions as well. While the first research question was a general one regarding looking into the effects of e-discussion on learners' FLA, the third research question investigated the effects of gender on FLA. Some items in the FLCAS were coded reversely as they were tackling FLA in a positive way,

unlike the rest of the items that describe learners' negative feelings (see Appendices B and C).

Finally, data gathered from interviews was qualitatively analyzed using a thematic line-by-line analysis and was categorized according to its themes to relate each theme with the research question it addresses, which is the second research question. The second research question deals with learners' awareness of their FLA and if the e-discussions affected them positively. The interview data was also transcribed and analyzed in Arabic, and then illustrative comments from the interviews were incorporated into the text after translating them into English.

37

Chapter IV. Results

The primary aim of this study is to examine the effect of e-discussions on adult Egyptian EFL learners' FLA levels. For this purpose, the researcher formulated three research questions, which were answered using a triangulation of instruments. The first research question addressed the effectiveness of e-discussions on learners' FLA levels. The second one investigated learners' perspectives of their FLA and of the e-discussion effectiveness. Finally, the third research question sought to determine if gender is a variable. The total number of actual participants in this study was 45 out of 49, with four participants excluded for various reasons. Both genders were part of the study; nonetheless, they were not equal in number. The total number of males was 35, whereas the total number of females was 10. All participants involved in the study attended both in-class discussions in addition to the e- discussion. Eight out of the 45 participants were interviewed afterwards.